This course gives a fundamental introduction to the field of deep learning. It will cover most relevant techniques to understand and model recent deep learning architectures. We will start with relevant basics, computational graphs, and the concept of error backpropagation. Then will we continue with shallow networks and most important optimization strategies. Afterwards, we will study several key aspects and architectural building blocks around convolutional neural networks, but also regularization techniques and how to build very deep networks. Moreover, we will investigate important generative models and other advanced deep learning approaches of relevance.